Many lines of evidence suggest that chronic exposure to hyperglycemia contributes importantly to the relentless decline in pancreatic islet [unreadable]-cell function characteristic of type 2 diabetes. The mechanisms of this phenomenon, termed glucose toxicity of the [unreadable]-cell, has been shown by us and by others in laboratory models of type 2 diabetes (animals, isolated islets, b-cell lines) to involve decreased levels of two proteins, PDX-1 and MafA, which are critical regulators of insulin promoter activity and thereby insulin gene expression. Our new studies are organized into 3 specific aims: Specific Aim #1. To determine whether beta cell-specific overexpression of glutathione peroxidase (GPx-1) will protect against worsening of insulin secretion and glucose tolerance characteristically observed in streptozotocin-induced hyperglycemia in C57BLKS mice, in high fat diet- induced hyperglycemia in C57BU6 mice, and spontaneously in db/db and NOD mice. Specific Aim #2. To determine whether islets isolated from pancreases of the transgenic mice studied in vivo in Specific Aim #1 are protected against abnormalities that are characteristically found with chronic exposure to hyperglycemia. These include (1) decreased expression of the insulin and PDX-1 genes and MafA protein; (2) decrements in insulin content and insulin secretion; and (3) increased apoptosis and decreased beta cell mass. Specific Aim #3. To determine whether stable reconstitution of MafA, PDX-1, or both in combination are sufficient to restore normal levels of endogenous insulin gene expression in glucotoxic beta cells. Experiments will involve physiologic studies in vivo and molecular biologic studies with isolated islets in vitro, as well as studies using retroviral overexpression of MafA and PDX-1. At the conclusion of these experiments we will ascertain whether enhancing endogenous antioxidant enzyme activity specifically in [unreadable]- cells of db/db and NOD mice will provide protective effects against the beta cell dysfunction and death that are characteristic of these animal models of spontaneous diabetes. We also will determine to what degree MafA, PDX-1, or both in combination play mechanistic roles in defective insulin gene expression in glucotoxic beta cells. Lay language: After onset of type 2 diabetes, high glucose levels over many years cause continuing damage to the cells that make insulin. This is thought to be due to oxidative stress because high glucose levels increase oxygen radical levels in these cells. At the time of onset of type 1 diabetes, cytokines, which cause oxidative stress, cause beta cell death. We will study whether increasing antioxidant protection protects these cells and prevents deterioration of glucose control in diabetes. [unreadable] [unreadable] [unreadable]